You are working with the text-only light edition of "H.Lohninger: Teach/Me Data Analysis, Springer-Verlag, Berlin-New York-Tokyo, 1999. ISBN 3-540-14743-8". Click here for further information.

	Index
See also: univariate statistics, data

Multivariate Statistics

Methods and algorithms using more than one input variable are usually called multivariate statistics. A special case with only two input variables is sometimes called bivariate statistics.

Here are some examples of problems which may require multivariate statistics:

• Can the risk of dying of cancer be related to the environmental conditions of the investigated persons? In order to find an answer to this question, one has to consider many factors, ranging from special chemical hazards to eating habits or smoking.
• Is there a way to identify counterfeit money? If the counterfeiting is badly done, a single variable will not be sufficient to identify forged banknotes. A set of variables will be necessary (size, color depth, geometry, ...).
• How is it possible to distinguish between different flavours of wine? The flavour of wines is determined by many parameters such as the concentration of various chemical substances, the temperature, the psychological conditions when drinking the wine, etc.

Last Update: 2004-Jul-03